Characterization and Mitigation of Impurities in Oligonucleotides Containing Methansulfonylphosphoramidate Linkages
Andrew Rodriguez, PhD, Director, Process Chemistry, Ionis Pharmaceuticals
Novel Oligo Linkage Tech Prompts New Impurity Control Strategy
Oligonucleotides containing mesyl phosphoramidate (MsPA) linkages can be more effective and safer to make than traditional therapeutic peptides. The only challenge, according to Ionis Pharmaceuticals, is the novel impurities these new linkage reactions introduce.
Modification is standard practice in the therapeutic oligonucleotide sector. The idea is that creating chemical bonds between the nucleotides of which the oligos are formed enhances their stability and makes them more active pharmacologically.
At present phosphodiester (PO) and phosphorothioate (PS) linkages, which protect therapeutic oligos from degradation, are the most widely used approaches. However, both increase the occurrence of nonspecific binding and can cause pro-inflammatory effects and damaging complement activation.
MsPA linkage is emerging as a potential alternative according to Andrew Rodriguez, Ph.D., director of process chemistry at Ionis Pharmaceuticals, who outlined some of the advantages at TIDES USA.
“Mesyl phosphoramidate is a relatively new internucleotide linkage with a bunch of promising properties. It enhances nuclease stability and protein binding properties of the oligos. It can reduce some of the immune stimulation and cytotoxicity. And it’s also supportive of the RNase H cleavage activity.
“The structure is very similar to a PO, but one of the non-bridging oxygens is replaced with a methane sulfonamide,” Rodriguez said, adding “Installing these links is actually quite easy. We install them using regular phosphoramidite solid-phase synthesis, but instead of using an oxidizer or sulfursing reagent after coupling, we use mesyl azide.
“Mesyl azide is able to undergo a Staudinger reaction with the phosphate, oxidizing it to the protective form of the MsPA linkage. After that, it’s fully compatible with all the other steps of all ago synthesis.”
Operator safety
Although MsPA linkages have some advantages over PO and PS links, there are difficulties. One major issue is the hazardous nature of the starting materials required, Rodriguez said.
“Mesyl azide is formed, most conveniently, from sodium hydroxide and mesyl chloride. The sodium azide, especially as you get into higher scales, becomes a significant toxicity concern. The azide itself — if it’s contacted with water or acids — can generate hydrazoic acid, which itself is a very big toxicity concern as well as an explosivity concern.”
“So all of these things present significant safety concerns for the operator.” However, Rodriguez continued, “We’ve done a lot of work to mitigate all of those concerns,” citing presentations from TIDES 2023 by colleagues Andrew McPherson and Francis Ring.
Impurities
The major challenge associated with MsPA linkage is the creation of impurities that have not previously been seen during oligo synthesis, Rodriguez said.
“When we look at oligonucleotides with mass spectrometry we’re used to seeing for impurities, the biggest of which would be the N minus 16, which is the PS to PO impurity. And then everything else is generally below 0.5%.
“However, we made the same sequence, but with four mesyl incorporations, rather than PS, we saw a number of impurities had gone up relative to the control. So these include things that are less than the main peak; those are minus 77 and minus 61, as well as things that are larger than the main peaks,” Rodriguez said.
Subsequent analysis allowed the Ionis team to develop mitigation strategies.
“Some of these impurities can be controlled, to a certain extent, by ensuring complete methylation during the methylation cycle. So we were able to reduce these to pretty much baseline level. So that left four impurities that remained unknown,” Rodriguez said.
These remaining impurities, which on the mass spec where plus 34 plus 35, plus 79 and plus 125, also need to be controlled. Fortunately, as Rodriguez explained, the team was able to quickly identify how the latter three impurities were formed.
“What we think is happening is that the guanosine is getting doubly isolated during a capping step and once it encounters mesyl chloride in the mesyl azide reagent it gets mesylated at O six. This persists through oligo synthesis until the treatment of the resin with dimethylamine, at which point dimethylamine displaces the mesylate. After that there are three potential pathways to follow one where you lose the acyl group, one where you lose the isobutyl group and one where you can lose both.”
This understanding was key to the development of a mitigation, Rodriguez said, explaining that they were “able to limit their formation by preparing the mesyl azide in a new way. So we prepare mesyl azide solution that is free of mesyl chloride by scavenging out the excess mesyl chloride, and we’re able to get down to about the level that we see in a typical PS oligo.”
In contrast, identifying the plus 34 impurity proved to be more of a challenge.
“Impurity levels in this case did not correspond with base composition, there wasn’t any one base that created this, it just happened regardless of the sequence. The level did increase with the amount of mesyl linkages, so it seemed to be something associated with the mesyl azide itself.”
And, when the team carried out a high-resolution mass spec analysis, they found that the impurity was associated with a chemical change – specifically a plus chlorine minus hydrogen transformation.
“This impurity was a starting material impurity, and we got to work trying to find a way to clean it out of our starting material. One of our vendors, Aurorium, was able to provide us a sample that was almost completely free of the chloro-mesyl, and when used tin oligo synthesis, this sample was pretty much able to eliminate the impurity from our system.”